Pavo Launches Instant Video Mode: 10-Second Clips, No Signup, Free Flash Model
机器之心 · wechat · 2026-09-07
Pavo, a video creation platform by AgnesAI, launched an instant mode: no signup required, one prompt or a first-frame image yields a 5-second video with audio in about 10 seconds, free three times daily. Machine Intelligence's hands-on test produced a text-to-video clip in 9.49s with stable characters and convincing rain/footstep audio; image-to-video also took 9s. The mode is powered by H3Max, fal's fine-tune of MiniMax H3 open weights.
After signup, users get free access to AgnesVideo2.5 Flash (720p, 4-12s, up to 5 images plus audio) and 200 daily credits (2 per second, 100s of video/day). The full AgnesVideo2.5 supports 720p/1K/2K and video references, ranks 19th on Artificial Analysis's text-to-video leaderboard at 1080 Elo, priced at $1.5/min — just 11 Elo below Veo 3.1's rank 14 at $24/min.
The platform also offers an infinite canvas workflow (node-based, regenerate from any intermediate step, parallel branches), reusable subject assets for character consistency, a 3D director stage for blocking and camera planning, an Agent that orchestrates full pipelines, and a dedicated short-drama mode. The article argues near-zero cost and instant generation turn trial-and-error into a frictionless habit, unlocking creativity.
More from Multimodal
- Anime-style TTS finetune Irodori-TTS-v4.1-Anime trends on Hugging Face — phasefield-audio · 2026-09-07
- Differentiable rendering going LLM-based makes sense; video generation may replace path tracing — jongranskog · 2026-09-07
- Astra generates a full 3D model and animation in Blender CLI from just 4 dog photos — Whattheduckying · 2026-09-07
- Midjourney v8.2 film-look portraits: the --exp parameter controls vintage grain — michaelrabone · 2026-09-07
- Risograph Minimalism Prompt Template for Minimal Two-Ink AI Illustrations — azed_ai · 2026-09-07
- Four models, one prompt: Astra, Kimi K3, Fable 5.1 and Sol compared on an underwater scene — heypearlai · 2026-09-07